Introduction to

Probability
and Statistics

Syllabus
Introduction to Probability and Statistics
Course ID:MA2203

Lecture-6
Course Teacher: Dr. Manas Ranjan Tripathy

Department of Mathematics
National Institute of Techonology, Rourkela

. . . . . . . . . . . . . . . . . . . .
1/9
Introduction to
Probability
and Statistics
Syllabus

Syllabus

Mid-Semester: Axiom of Probability(Motivation, Various types

of definitions), Some basic results on probability (Boole’s and
Bonferoni’s inequalities), Conditional probability, Bayes the-
orem, Probability for independent events, Random variable,
types of random variables(discrete and continuous), Cumu-
lative distribution function, probability mass function, prob-
ability density function, Mean, Variance, Standard deviation,
Moments (central and about origin), Moment generating func-
tion, Special types of random variables, discrete: uniform,
binomial, geometric, Poisson, Hyper geometric, Continuous:
Uniform, Normal, Gamma, Exponential (one parameter).

. . . . . . . . . . . . . . . . . . . .
2/9
Introduction to
Probability
and Statistics

Syllabus
After Mid-Semester: Two dimensional random variable, joint
CDF, joint PDF/PMF, Marginal distribution, Conditional distri-
bution, Calculating probabilities using two dimensional ran-
dom variable, Sampling (with and without replacement), Dis-
tribution of sample mean and variance in the case of nor-
mal distribution, Estimation: Point estimation, Method of mo-
ments, method of maximum likelihood, Confidence Intervals
for mean and variance in the case of normal, Testing of hy-
pothesis (parameters of normal distribution), Goodness of fit
Chi-square test, Regression and Correlation analysis, Rank
correlation coefficient.

. . . . . . . . . . . . . . . . . . . .
3/9
Introduction to
Probability
and Statistics Moment Generating Function(MGF): Let X be a random variable de-
fined on a sample space S. We denote the MGF of X having some
distribution as MX (t) and is defined as
Syllabus
∑ tx
MX (t) = E(etX ) = e j P(X = xj ), if X is Discrete
j
∫ ∞
= etx f (x)dx, if X is Continuous.
−∞

( t 2X 2 tr X r )
MX (t) = E(etX ) = E 1 + tX + + ··· + + ...
2! r!
t2 tr
= 1 + E(X ) + E(X ) + · · · + E(X r ) + . . .
2
2! r!
∑∞
tr ′
= µr ,
r!
r =0

where µ′r = E(X ) is the r th moment about origin of X .

r

tr
That is, we see that the coefficient of r!
in MX (t) gives the r th moment
about origin that is µ′r .

. . . . . . . . . . . . . . . . . . . .
4/9
Introduction to
Probability
and Statistics
Further differentiating MX (t) with respect to t r times and putting
t = 0, we get
Syllabus
dr µ′ t2
r
[MX (t)]t=0 = [ r r ! + µ′r +1 t + µ′r +2 + . . . ]t=0 = µ′r .
dt r! 2!
The moment generating function of X about any point ‘a’ is given by

M(X =a) (t) = E(et(X −a) )

t2
= E[1 + t(X − a) + (X − a)2 + . . .
2!
tr
+ (X − a)r + . . . ].
2!
Since it generates the moments, it is known as the moment gener-
ating function.
The moment generating function if exists is unique. That is for a
given random variable X , we can find its moment generating func-
tion uniquely and vice versa. In another word, we can say that the
moment generating function characterizes the distribution function.

. . . . . . . . . . . . . . . . . . . .
5/9
Introduction to Relation between Moment about origin (µ′r ) and Central Moments(µr ):
Probability
and Statistics µk = E(X − µ)k = C(k , 0)µ′k − C(k , 1)µµ′k −1
+C(k , 2)µ2 µ′k −2 − · · · + (−1)k µk .
Syllabus

µ′k = E(X k ) = E(X − µ + µ)k = C(k , 0)µk + C(k , 1)µµk −1

+C(k , 2)µ2 µk −2 + · · · + µk .
Exercises:(i) Let the random variable X has the pmf P(X = r ) =
q r −1 p; r = 1, 2, 3 . . . Find the mgf of X and hence its mean and
variance. 0 < p, q < 1, p + q = 1.
Ans:
∑∞
MX (t) = E(etX ) = etr q r −1 p
r =1

p∑
∞
= (qet )r
q
r =1

p t∑
∞
= qe (qet )r −1
q
r =1
( pet )
= pet [1 + qet + (qet )2 + . . . ] = .
1 − qet
. . . . . . . . . . . . . . . . . . . .
6/9
Introduction to (ii) The probability density function of the random variable X is given
Probability
and Statistics by
1 ( |x − θ| )
f (x) = exp − , −∞ < x < ∞.
Syllabus 2θ θ
Finf the moment generating function of X , and hence mean and vari-
ance.
Ans:
∫ ∞ ( |x − θ| )
1
MX (t) = E(etX ) = exp − etx dx
−∞ 2θ θ
∫ θ ( (θ − x) )
1
= exp − etx dx
−∞ 2θ θ
∫ ∞ ( (x − θ) )
1
+ exp − etx dx
θ 2θ θ
∫ θ
1 1
= exp{x(t + )}dx
2θe −∞ θ
∫ ∞
e 1
+ exp{−x( − t)}dx
2θ θ θ
eθt eθt
= +
2(θt + 1) 2(1 − θt)
eθt
= = eθt (1 − θ2 t 2 )−1
1 .− .θ2. t .2 . . . . . . . . . . . . . . . .
7/9
Introduction to
Probability Hence
and Statistics

θ2 t 2
MX (t) = eθt (1 − θ2 t 2 ) = (1 + θt + + . . . )(1 + θ2 t 2 + θ4 t 4 + . . .
Syllabus 2!
3t 2 θ2
= 1 + θt + + ...
2!
Mean of X is the coefficient of t in MX (t) which is θ. That is µ′1 = θ,
2
µ′2 = coefficient of t2 in MX (t) which is 3θ2 . Hence variance of X ,
V (X ) = µ′2 − (µ′1 )2 = 3θ2 − θ2 = 2θ2 .
(iii) If the moments of a random variable X are given by E(X r ) = 0.6;
r = 1, 2, 3, . . . show that P(X = 0) = 0.4, P(X = 1) = 0.6, P(X ≥
2) = 0.
Ans: The moment generating function is given by
∑
∞
tr ′ ∑
∞
tr
MX (t) = µr = 1 + (0.6)
r! r!
r =0 r =1
∑
∞
tr ∑
∞
tr
= 0.4 + 0.6 + (0.6) = 0.4 + 0.6[ ]
r! r!
r =1 r =0

= 0.4 + 0.6et .

. . . . . . . . . . . . . . . . . . . .
8/9
Introduction to
Probability
and Statistics

Syllabus

But we have,
∑
∞
MX (t) = E(etX ) = etx P(X = x)
x=0
∑
∞
= P(X = 0) + et P(X = 1) + etx P(X = 2)
x=2

Comparing these two, we get the required result that is P(X = 0) =

0.4, P(X = 1) = 0.6 and P(X ≥ 2) = 0.

. . . . . . . . . . . . . . . . . . . .
9/9